In my earlier post I talked about why you should use a KPI tree. If you were convinced of their value, here's an example of how to build one...
Build out the branches of a KPI Tree
Below is a single branch from the earlier example, showing how you move progressively closer to something you can directly measure as you move towards the ‘measures’ level of the diagram.
Drilling down one branch of the KPI Tree
As you go down the levels there is a one-to-many relationship. So, for example, our strategic objective ‘Be healthy’ has several enablers that link into it. Here are the enablers I’ve identified:
- Eat well
- Control/lose weight
- Be aerobically fit
- Sleep well
Each of these will have several tactical enablers living below them, and measures below them. Here I have fleshed out the Eat well branch:
Expanded branch of the KPI Tree
Any ‘node’ on the tree can be linked to any other to show a relationship. Use colour, intensity or terminator shape to distinguish between the three link types, described below.
Link type 1 - Cause and effect
Where one activity directly influences another. This the most common type of relationship, so I use a plain grey line for this.
Examples of cause-effect relationships include:
- ‘Sell popular flavours’ causes ‘Increase in ice cream sales’
- ‘Consumption of free salty snacks’ causes ‘Increase in drinks sales’
Link type 2 - Conflict
Where one activity conflicts with another. I use a red line to show this.
Examples of conflict relationships include:
- Minimise 'Average Handling Time’ conflicts with ‘Maximise First Touch Resolution’.
- ‘Minimise performance rewards’ conflicts with ‘We have motivated staff‘.
Link type 3 - Companion
Where one measure is a subset of the other, or there is significant overlap. I use a double line for this - implying a two-way relationship.
Examples of companion relationships are:
- 'Weight' is a companion measure to 'Body Mass Index'
- ‘Reduce loss from car accidents’ is a companion objective to ‘Reduce injury from car accidents’.